The large scale structure of the Internet

Abstract

The Internet is a prototypical example of info-structure that has grown following a self-organized dynamics. Even though governments are waking up to the reality that the Internet is a critical infrastructure that has dramatically changed our way to access information, exploit social relations, and run commerce, nobody has ever run the Internet growth or drawn a blueprint for its development. The dynamics of Internet is indeed defined by the interplay between cooperation (the network has to work efficiently) and competition (providers wish to earn money). These evolutionary prin- ciples have shaped an intrinsically heterogenous system ruled by different administrative policies whose complicate structure cannot be found in any repository or “official” map. For these reasons, in the last years, several research groups have started to deploy technologies and infrastructures in order to obtain a more global picture of the Internet. Several studies aiming at tracking and visualizing the Internet large scale topology and/or performance are now providing Internet maps at different resolution scales. These projects typically collect data on Internet nodes (routers, domains, etc.) and links in order to create a graph-like representation of large parts of the Internet. The obtained Internet graphs exhibit most of the features characterizing large scale complex networks and in the present chapter we provide a review of the results obtained in the characterization of their structure. We also present a critical discussion of the eventual experimental biases that might lead to erroneous conclusions on the actual topological properties of the Internet network. Finally we discuss some novel measurements aimed at uncovering the hierarchical and ordering principles underlying the Internet structure.